Lenovo

Building on our breakthrough performance in high-performance computing (HPC) and our expanding global HPC customer base, Lenovo is enhancing our global HPC/AI team with specialist performance engineers. We aim to lead the next generation of air and Neptune liquid-cooled high-performance computational platforms. To achieve this mission, we are looking for a motivated and innovative HPC/AI Performance Engineer to join our team. This role involves rigorous testing of our systems at scale to ensure optimal performance helping us to build some of the world’s most impactful HPC systems to help solve humanity’s greatest challenges.

You will report to the HPC Performance and Benchmarking team, supporting customers worldwide. Your role will involve participating in both virtual and in-person meetings with team members, sellers, architects, and customers. Occasionally, you may need to attend customer sites.

What You will do:

• Execute HPC/AI benchmarks using Lenovo technologies and prepare reports related to customer RFPs.
• Investigate and resolve computational performance issues related to HPC/AL applications and Lenovo servers.
• Support delivery of Lenovo Solutions, including reference architectures and validated designs.
• Analyse and characterise scientific codes and build performance extrapolation to future generation of HPC/AI hardware.
• Interact with customers and the Lenovo sales team to offer insight into workload performance characteristics that drive system configurations.
• Complete competitive comparison studies of different technologies to showcase Intel technology advantages.
• Develop seller enablement collateral for Lenovo Sellers and Business Partners, and participate in sales enablement activities.
• Attend conferences and workshops relevant to HPC and stay current with HPC trends.

What you bring:

• Bachelor’s degree in Computer Science or relevant experience. Candidates from scientific backgrounds such as physics, maths and other analytical sciences are welcomed, as well as someone who just completed their PhD degree.
• 3+ years of experience working with Linux; with scientific code and performance tuning.
• 2+ years of experience of scripting and coding, with specific experience of more than one of: bash, python, C++, C, Fortran.
• 1+ years of experience with High Performance Computing or Artificial Intelligence technology, including servers, Infiniband interconnects, high performance Ethernet and accelerators.
• 1+ years of experience of more than one of: HPL, Stream, IMB/OSU, TensorFlow, PyTorch, MLperf.
• 1+ years of experience of using parallel and/or accelerated applications using more than one of OneAPI, Intel Gaudi software suite, OpenMP, MPI, CUDA, ROCm, OpenCL, SYCL.

We will also consider candidates who lack some of the requirements but can demonstrate ability to learn quickly.

The following competencies are welcome:

• Experience of running performance tests and benchmarking codes.
• Experience of using or developing parallel and/or accelerated applications using more than one of OpenMP, MPI, CUDA, ROCm, OpenCL, SYCL paradigms.
• Experience of production HPC environment: large-scale filesystems (ideally Storage Scale), batch scheduling (ideally SLURM) as well as common HPC SW and management tools.
• Experience with analysis and profiling tools for HPC/AI codes: Intel OneAPI suite (Vtune…), AMD (uProf), nVidia toolkit.
• HPC application characterization.
• Performance optimization of HPC or AI codes.
• Experience with DevOps automation, configuration management and CI/CD.
• Knowledge and experience working with container-based applications.
• Project management notion.